If the device has balanced connections, that's what Amir will use. If not, he would say so and tell you the reason. I have no doubt he measured the JBL using balanced, and that's why he measured it at 4 V. So the comparisons would be fair.
Again, 2 V unbalanced = 4 V balanced, because +2V minus - 2V = 4 V If anything, theoretically speaking, the Denon would have been in disadvantage because it does not come with balanced connections. Balanced should have lower noise, even if the device is not (more often not) truly end to end balance/differential. Fully end to end balance/differential connections such as the Denon AVP-A1HD1 that was the world's first and only (may still be??) at the time, would have lower distortions as well because if perfectly designed and implemented, even order harmonics would be cancelled.
Just a note, in practice, I have never seen any bench measurements that show noticeably lower distortions or even better SNR based on S&V'e measurements, in fact, more often than not, they showed a little better performance using unbalanced/RCA's. If I remember right, HTHF's did show better measurements with balanced, though still nothing significant.
