Very cool. Thanks, guys.
An update from my end:
The labelling for Fig. EC-20 is, indeed, reversed re Items 7 & 8.
The 'case cover' is simply threaded onto the main casing, so it's easy to get off (big hex-nut fitting). After the cover is off, though, there's a brass cap (from which a central shaft with the adjusting nut protrudes) that looks like it's a press-fit into the main casing... so I'm probably never going to get to inspect the innards without destroying something.
I've got my thermostat soaking in CLR now. There was a lot of 'action' right off the bat, so my suspicion about mineral build-up might be correct. We'll have to wait to see, though, whether it frees up the workings.
Interesting that the designers decided to add an adjustment facility for such a simple device. That may explain, too, why there's a threaded cover rather than just making the device completely sealed. Maybe they knew something we don't !
Don't know if anyone else spotted this, but the intended flow direction for the coolant is the reverse of what I would have expected. According to the FSM diagram (unless this is another labeling error), when the T-valve is open, the coolant is flowing out of the rear bypass tube, through the T-valve, then into the manifolds, and finally exiting into the main thermostat casing at the front of the engine. I always thought the flow direction was the opposite (i.e. from front to back, and then around the rear of the engine into the Y-fitting). I guess I'll have to go back and look at those coolant flow diagrams that someone came up with in the midst of that long debate over whether or not it's safe to bypass the cabin heat coolant circuit.
I'll report back tomorrow on whether or not the CLR soak worked.