As is almost always the case, "it depends".
However, I can certainly make that argument that typically, assuming a limited budget, bookshelves are the way to go. Here are my experiences/thoughts on this:
Tower speakers involve larger panels that
are prone to resonance. Simply put, a 6" cube made of 3/4" mdf will pass the knock test quite easily, but a 2 foot cube made of 3/4" mdf will not. The solution is to compensate with bracing to the larger cabinet, but this adds cost to construction as will as shipping and handling (increased weight).
Understand that this resonance doesn't necessarily sound bad. My favorite example is comparing Hotel California where the resonance I heard made for some very full and rich bass that was lovely vs. about anything by Steely Dan where being tight and precise is king! I think, like me, the vast majority of people would prefer the resonant cabinets for Hotel California and the non-resonant cabinet for Steely Dan, assuming the other attributes of the speakers were "equal" (and the resonances were reasonably tame).
I expect the best ($$$) offerings of Revel, Focal, B&W, etc. to have the required bracing to eliminate resonances to be on par with the same series bookshelves, but look how much you pay for it! As a general rule of thumb, I expect there to be audible resonance from any tower with a street-price under $1000 each.
The premium Klipsch speakers I know of do not offer a Bookshelf equivalent!
Totem makes a thing of using natural resonance as part of their sound. I found them to be very nice for something like an intimate trio, but felt they fell apart when subjected to more complex music with lots of syncopation (especially as the volume went up).
With speakers like the Klipsch you mention,
where the drivers are identical for the towers but an extra woofer, I would expect very similar sound character until you got into lower frequencies. If you crossed them at 80Hz instead of 60, I would feel pretty comfortable believing you would not hear much difference aside from resonance. However, at 60Hz, I could see the inherent roll-off of the single 5.25" driver interfering with the rate of the AVR's roll-off. SVS recommends using an AVR to cross these speakers to sub at 80Hz. If you do not have an AVR and have to cross to the natural roll-off of the speaker, they recommend setting the sub XO to 60Hz. So if you were to XO at 60Hz using an AVR, you would essentially have a compound roll-off of the 150M's - both the AVR and the speaker together cause a steeper roll-off than you would want.
It is unfortunate that your 150 and 250 are not from the exact same series, but I would suggest you do an A-B comparison of these two speakers in whichever room you are considering upgrading to towers. Use the receiver to roll off both speakers at 80Hz and just see what difference you get. I assume you have two receivers, so I would suggest you use both so you can level match and instantly switch between the two pairs of speakers. A source such as a DVD/BD player usually has two outputs that are both active (my Oppo has two HDMI's that are active, but even my CD player has a toslink and RCA outs that are both active. Last, IME, since you are comparing pretty equivalent speakers, once you reach some conclusions and repeatable observations, you need to swap the speaker locations, as that is a very likely to contribute to what you hear.
With speakers that involve different drivers to increase the bass performance, you are much more likely to realize a different sound. Let me use the SVS Ultra Towers vs Bookshelves as an example. The towers take the same tweeter and midrange from the bookshelf and add a second 6.5" mid-woofer and two 8" woofers to get f3 of 28Hz. A key point is the mid to woofer OX is 160Hz in the towers while the bookshelf has an F3 at 43Hz...so the single driver in the bookshelf is expected to play down to 43Hz, while SVS felt it was better to cross to the 8"woofers at 160Hz! Now, I can tell you from personal experience that the Ultra Bookshelves do not have noticeably sloppy bass, but were I to A-B the towers against the bookshelves (with an 80Hz XO so the differences in lowest bass are avoided), I would expect there to be some readily audible differences in the frequencies below 160Hz.
So now, I think we can assume that the towers will sound a little different and presumably better than the bookshelves (at least in this low frequency range), and it becomes a question of whether the difference in price ($1000/pr vs $2000/pr) is justified.
If your budget is $2000, I think it is a no-brainer...Bookshelves plus $1000 of subwooferage will outperform towers. Furthermore, we might look for a sub known to offer higher performance at higher frequencies with the option of XO at 100 or 120Hz to relieve the 6.5" mid-woofs of the most demanding lower frequencies. However, if you have the money to get the Towers without it coming out of your subwoofer budget, I do believe the Towers will get you some audible improvement.
If I can use
@Pogre as an example (which is poignant because of the irony), he bought the Ultra Towers and reveled in the solid bass (as you might expect from SVS) that they provided. However, after he went through the process of measuring and miniDSP'ing his system, there was no way around it...he (perversely) needed to cross these towers to the subs at 100Hz for the best sound quality. Much of that has to do with the room and the restrictions on where you can position tweeters to get proper sound from your mains, so someone else might have a different outcome. Obviously, he spent a lot of money for performance into the 20Hz territory that he will likely never use! With the benefit of 20/20 hindsight, and if we just look at the stereo system, he ideally may have done better buying a pair of Philharmonitor Audio BMR speakers* for $1350 instead of the towers (but they would not have matched the rest of his system so well as staying with the Ultras all around and I doubt he has any serious regrets).
*Technically, the BMR is a bookshelf, but at 20" tall is is really something of a hybrid to provide deeper bass than a normal bookshelf would!
In any case, I generally consider Bookshelves with subs a better deal.
