Kind of a necrobump but I recently came across this thread. Since this was my design I figured I'd chime in.
The input buffer convergently ends up resmbling the Cornish buffer, though this circuit topology was not invented by Cornish. You see bootstrapping in a wide variety of high impedance scenarios. You can even do it with an op amp instead of a BJT and obtain astronomically high input impedance (tens of giga-ohms range), which is great for things like scientific instrumentation amplifiers.
As for IC1.2, it is arguably a relic of the dev process. The prototype provided "handles" for a structured mid boost which could have been more intense but ended up being fairly subtle. Ultimately it provided some much-needed polish. Could it have been done more simply? Probably. But a few cents of extra resistors/caps is cheaper than a new PCB layout.
Putting a cap after a voltage reference buffer does essentially nothing except force the op amp to drive a capacitive load, which most op amps hate doing. It's a common mistake, classic symptom of the frequent cargo cult-ing of pedals. There are ways to convert that op amp buffer into a LPF which effectively increases the regulation of the VREF node (there are some Burr Brown app notes about this) but it's generally not necessary.
Not quite accurate about that series potentiometer. It doesn't keep the op amp from distorting, it sets the gain. The feedback diodes keep the op amp from distorting. Important distinction.
Also, re: the 4 years thing... factor in a pandemic, a collaborator with his own business to run (
), and a couple tours for Matt's band, and suddenly you blink and wonder where the time went.