Interesting. And industrious on your part, too.
One of the aspects I tend to ramble on about, whenever the topic of emulating reverb comes up is that, in the real world, earliest reflections have more high-frequency energy than later reflections, since the hi-freq stuff gets "eaten up" by imperfectly-reflective surfaces. What that suggests is that if multiple delay paths are employed, the path/s used for later relections should employ more low-pass filtering, or filtering at lower corner frequencies, to mimic that quality. Personally, I also find that adding a bit more lowpass filtering to feedback paths (making each subsequent repeat have juuuussst a little less top end than the previous repeat) also works toward that end.
Nne of that is to suggest that omitting such filtering isn't musically interesting or useful. But if the goal is to achieve something closer to natural reverberant spaces, then one needs to do what those spaces do.
Apart from that, keep up the good work!
I agree about the increased low-pass filtering on repeats. I've been playing with some form or another of this circuit for about a month, 4 or 5 hours per week (maybe more) trying different configurations. I found that coming off of pin 14 (like the standard, good sounding, delays), has a bit more filtering at the expense of cutting out the main signal to the following delay. I tried to compensate by adding clean signal directly to the 3rd chip's input, but that also sounded unnatural because there would be some repeats with more sharpness than others. One thing I like about this setup is that the original signal goes through the fairly hefty LP filter in the 15-16 opamp before leaving to go to the 3rd chip. The original signal is filtered once in the first stage of parallel chips, then it's repeats are further filtered in the 3rd stage. The repeats of the first set of stages are still re-filtered on each repeat.
While typing and thinking right now, I think it may be worthwhile to have a switch for each section to select between the pin-15 output, or the (standard) pin-14 output, which has two more filter stages, like the right-hand chip in the schematic.
I'm still quite happy with the sound, it's very close to what I had in my head before the experimentation began. I wanted a reverb that sounded a little more like delay, but with the option to sound much more like delay.
**edit: it's obvious in the video, but the phone recording picked up quite a bit of direct string sound (especially in your right ear), while at the same time filtering out low frequency content. In the near future, I'll make a proper recording with an SM57.