Well, I'm back and the news isn't good. I removed the valve and union coupling and soldered on a couple of threaded ends. I put an analog WATT water pressure gage on each end (in turn) and let it sit for 10 minutes before taking the measurement. I measured flow rate (quantity of water into buckets at full throttle over one minute). I also measured the temperature of the water in the buckets. Note that I didn't start the clock on the hot water until the cold water had made its way through (i.e., we were at a steady state). The results:
cold side: 77 psi, 6 gpm, 54 degrees
hot side: 77 psi, 2.9 gpm, 144 degrees
I'm at a loss. I was hoping for there to be a large pressure difference. But the pressures are as close to equal as I can read on the gage. The water heater is working almost exactly to spec (flow rate at a temperature increase).
I originally was thinking that I would open up the y-check valves, but they're pretty tight so I just hit them a couple of times with the wrench

. They seem to be working properly based on the flow at the drain valve in various configurations.
The only thing that I can think of that's left is to increase the distance from the mixed outlet to the first bend (currently just 2"). Or get a third valve.