It uses silicon PNP transistors, but is still a negative ground effect (so daisy chaining power supply is still ok). I guess Sean had a bunch of extra 2N3906s lying around. I took a few liberties with the power filtering and polarity protection, but nothing that will effect the original sound. The yellow/ceramic 100n cap going from 9v to ground is optional, as it's not all that necessary, but is in the original. It's setup for board mounted pots if you so desire.
This fuzz effect is essentially a Fuzz Face with an emitter follower up front and clipping diodes on the output. It produces a treble-heavy fuzz with a good bit of volume. Original transistors were BC238B for Q1-2 and a BC239C for Q3. Some schematics floating around call for 2N5088s, but shoot for transistors in the 200hFE range (2N2222, 2N3904, etc).
Q1 is a KSP42, which seems to be pretty similar (if not the same) to a MPSA42. I've read of MPSA18s working in that spot as well. Q2 is a 2N4401, which should be easier to track down, though a 2N3904 should work just fine there as well. As always, socket and experiment. The depth pot is listed at 470k, but a 500k will work just fine.
This is based on Tim Escobedo's LoFoMoFo, but tweaked for more bass frequencies. This circuit was actually my first build and it sounds great. Although, I made the n00b mistake of using a 10n cap on the input instead of the 100n because I misread the capacitor code. So I guess I should say it sounds good with a 10n input cap. Can't really speak to the stock sound, but it's apparently very fat.
The circuit runs off of +1.5V, just like the FZ-1A. The problem with this is that we would need either a dedicated power supply or run it off a AA battery, and you wouldn't be able to wire up an indicator LED. So what I've done is made a board for the original, and did some digging to see how to get +1.5V from a +9V source.