Hi Antti,
Wow, you're really digging in, asking for the secret sauce! I'll do my best to answer your questions.
1. DualVerb has 2 stereo reverbs, so the FXmix knob mixes the input into these reverbs in stereo, a PRE reverb mixer for each stereo side basically. So, A 10:B 10 means they are fed with equal L and R input signals. If you're hooked up with Mono In, it just routes and mixes the one input to both reverbs.
Furthermore, when the pedal is hooked up Mono In Stereo Out, you have the DUAL MONO option on the FXmix, this will send the input to both reverbs, but the Left output will be all A reverb and the Right Output will be all B reverb. If the pedal is hooked up Stereo in Stereo Out, the DUAL MONO option sets reverb A and B to process the L and R inputs completely independently and send them to their respective outputs.
Of course, this gets a bit more in depth with FREEZE. When frozen, the FXMix knob goes POST and mixes the OUTPUT of the two stereo reverbs.
2. ModEchoVerb is routed Modulation -> Reverb (modulated) -> Feedback Delay. The output, however is the parallel mix of all 3 of these, so imagine, that even though their input is fed as described above, we get a combined output by tapping off of each one. The FlangeMix and ChorsMix, not only control the Mix in the parallel output, but also, to some degree, the Depth of the Chorus or Flange in the Modulation Block. You're right, the SweptVerb turns off the Modulation Block and modulates the Reverb itself. It's a completely different reverb structure from the other algorithms, and the modulation is a bit more predictable. I've achieved very similar sounds to a modulated delay feeding a reverb in ModEchoVerb, but the structure is technically different than that.
3 and 4. Shimmer is structured as L+R input summed with L/R feedback, summed to mono-> mono in, stereo out reverb -> (I'll just write the 1 feedback path, but there are 2) Delay ->PitchShifters -> back to the beginning to sum with input. The output(s) are taken after the reverb, so you can dial all the pitch out. The delay really functions to stretch out the swells, it can open up the shimmer and give it breathing room.
Pitch Dcy is a macro control, it controls feedback around the entire structure. The other Hi/Mid/Low Dcys allow freq band based feedback control. The idea being you can target certain "colors" in the pitch swells or dial down "colors" you don't like for certain input material.
Does this help?
