I fully understand the pain.. I've always said it, always been a foot behind when it came to L2.
This didn't stop me from getting into Lightning, running nodes and joining in the fun of experimenting with channels (when the fees weren't this high).
But Bitcoin's whitepaper, coupled with the accessibility of running verification plug&play nodes, and the investment/PoW necessity for mining is a simple game overall.
Lightning has a scalability problem, and it not one down to the tech, but to its accessibility. I think developers like nostr:npub1xnf02f60r9v0e5kty33a404dm79zr7z2eepyrk5gsq3m7pwvsz2sazlpr5 have been monumental over the past year showing an embedded node is well possible. Whether it's feasible is something that's currently being battle-tested.
I've always thought of Bitcoin nodes running in households like people buy Wifi Routers, this seems really feasible to me. But having to deal with channels is a whole other game.
Personally, for an L2 proponent to succeed, it needs to find a way to be as straight-forward as MAC addresses and Ethernet frames are talking to both L1 and whatever else.