Atmasphere, unfortunately the way you present the concept of feedback with the term "propagation delay" helps to 'propagate' (sorry, couldn't help it ... :-) ) a misconception about negative feedback. Yes, "negative feedback is always lagging behind" but it is still able to "correct the signal it was supposed to". If it ends up lagging by 180 degrees, half a wavelength, with gain still happening, then you have positive feedback, and an oscillator, and fried tweeters!
So, the feedback ALWAYS lags, even by some tiny amount, but that is just part of the understood design of any working circuit. You can easily buy one of those "terrible" opamps with much more global feedback then a typical audio power amp, that is working comfortably at microwave frequencies, that is, beyond 1,000,000 kHz -- not bad for such a "mediocre" part!
Global feedback in an audio device, on the other hand, with discrete components is typically rolled off (because it has to be for stability, remember the oscillator!) to become non-existent at the higher frequencies, where it is most necessary, and where the ear is most sensitive to distortion -- that is part of the real problem. Also, that is why nearly all gear has rising distortion at higher power, and higher frequencies; in other words, it is not the presence of feedback that's the problem but that at the very time it is most needed, it's done a bunk ... :-) !!
Frank