... Will two ST heads "lock" to each other if operated in parallel ?
You better hope so.
Like you, I spent a lot of time researching this a few years back. I decided against it.
Main reason was the Listeroids have a lot of cyclical variation in the speed throughout the 4-stroke cycle. If you have one generator by itself, the only problem is typically some flicker. If you have two in parallel, circulating currents will flow between the two generators if the power strokes are not in exact lock-step. My plan was to oversize the generator head so that they could handle it, but I really wasn't able to calculate the magnitude of the problem. It was going to be a big investment in time and money, and I finally judged it to be too risky.
AFAIK, it's not been pulled off...
I also thought about the induction generator idea, but the more I read about them, the less I liked them. Plus, you don't end up with a truly redundant system. The induction machine is pretty much useless by itself.
So, yes, it pegs the cool-O-meter, but the practical-O-meter pretty much reads zero...