That's what I thought, I think the less hurdles between one subwoofer and another, the better the results should be, but I'm still wondering if this would cause cancellations between them = /
No. You are misunderstanding. The low frequency waves are omni-directional. They don't care what is in your room. (For example, this is why down-firing subs still exist: they work just fine aimed at the floor. I've heard of another guy that turned one of his subs to fire straight into a wall...) Diffraction is much more of a concern for your Center: Mid-Woofers and Tweeters are going to be affected much more significantly.
Your concern with Subwoofers should be at learning about your room resonances, and the Modes you will have to work with. Placing your subs next to a tower that is generating some lower frequencies is more likely to create a cancelation. Having your subs equally spaced so that the soundwaves bounce back and interact with each other can be just as destructive. The idea of spreading multiple subs around is to excite as many of those room modes as possible, thus evening out the bass response throughout more of the room.
Multiple Subwoofer after Geddes
http://seriousaudioblog.blogspot.com/2012/05/two-great-articles-on-multiple.html
You don't have to do this to get good results, however, I have found in my room, the front wall with almost full-range towers, and corner loading to be pointless. I used the subwoofer crawl to find good placement location, then I put my subs there and tweeked them slightly. I have not done much in the way of room measuring since, but my Bass is much better than by the more traditional/dogmatic approaches.
So to answer your repeated question. Neither of those is necessarily correct. Without measurements to show how they perform, its just a photo. I would not pattern my room after those.