I like your ideas, I've been optimizing TTLL for 2 years now and I'm working on my version 5 alg spreadsheet, I think there are some good stuff in it. Tripod for CFOP or ZZ, also is this F2L insertions preserving the tripod or is this a LL thing with OLL's and COLL's involving the tripod?
OLL plus a flipped F2L edge has 57 OLL's x 4 possible F2L unoriented edges x 4 becuase every unoriented edge must be in a pair so that influences the OLL = 912 algs, I don't think that would be something many people would learn IMO.
I like the ZBLL idea the most because it is something that will benefit many people, applies to many methods, and is developed enough to have some of the heavy lifting done for me such as images for all the cases. I think I will start with the ZZLL's since it is an algset I've tried learning but the algs were not that great when I tried and maybe move on to ZBLL from there.
HKPLL I agree with what you're saying on that topic. I dont think I'll be generating HKPLL soon as there is nothing to go on and I'm not that good with cube thoery to come up with all the possible permutations of edges, so if someone had images for all the cases I would be more inclined to do that but as of right now I've searched and come up with nowhere to start from. So yeah maybe in a couple years I'll google search it, see what people have about it.