>
The op-amps needing replacement are LF356n and LM301an.Beware the 301! It is very versatile and was used in several wonderful ways that do NOT match "modern" chips. In particular, several places use "feedforward compensation" (150pFd cap from -in to Comp) which makes the 301 incredibly fast, and very nearly single-stage. If a 301 has anything more than a 30pFd Comp cap, you need to re-think what Ward-Beck did. Or just leave it alone.
An issue: plans show +/-21V on LM301an. This is over the usual spec for 301 (36V max). There may be some special-spec versions rated 44V. I assume either it isn't really +/-21V or they got 44V rated parts, because 20 power-on years would have weeded-out the weaklings.
This business is clearly tacked-on, probably to control supersonics.

After you fix your oscillations (there is no such thing as "a little oscillation", oscillation is always bad and usually unpredictable), if you are going into modern recording gear you do not want this low-pass filter. (It reduced problems with old recorders that never expected to see anything over 20KHz.) Just do the red-line modification.
All caps look more than big enough to me, even over-sized for electrolytic distortion reduction. Bandwidth would extend to the geological, except the 70/100/150 high-pass seems to be 10Hz (12dB/Oct) when not-engaged. And that is 0.22 0.1 film caps, no gross distortion.
It is a pretty complicated path with a lot of push-pull emitter followers. Yet most of them work in Class A, and moreover are buffering BiFET amp chips which suck a lot less if buffered. And the "rude" way they bias the emitter follower forces the BiFETs into Class A. This isn't going to sound like the usual box of TL074s.
I suspect it will be the essential dead-clean 1980 console. Perhaps a little soul-less, but to "give it flavor" probably means messing it up. Go get an Altec Tube box for "color".
Sadly I do agree that the tantalums should probably all be replaced. They were hyped as "long life" but that has not turned out to be true.