S-Man, you have my sympathy and respect. Us analog blokes have been hearing about the glories of single-ended this and that, discrete this and that, hand-selected capacitors that cost a fortune, zero feedback, etc etc for decades. Turns out they are pretty much "old wives tales", passed on from one generation to the next. Yes, the original 741 and TL074 op amps were junk for audio, but once some smart guys at Burr-Brown started working on the problem, things got a whole lot better, even though the rhetoric didn't. People who have heard an OPA627 know what I'm talking about. But the newer opamps are even better.
However, all new op amps are not equal. I listened to at least 3 or 4 different pairs before deciding on the TI1611 and TI1641 (bipolar input and FET input, respectively), also designed by the Burr-Brown guys - now working for TI. If you listen with your ears rather than your pre-determined opinion, you can make some real progress in SQ. From my experience, the amount of "control" you have over the signal makes the difference, and feedback circuits using op amps fill that bill.
I also design guitar amps with single-ended tubes (valves) and distortion that measure up to double digits, so I know what that sounds like. There's nothing like it for generating that spine-tingling lead guitar riff. But when you want to reproduce that from a recording in your living room, it takes a different kind of system to do that faithfully and with stage-level dynamics. As I replace the discretes in my earlier circuits with op amps, it just sounds better and better to me.