Plan B will be a slower increase in hashrate after the next halving, that's all. That's not such a bad thing: it will still be improssible to do a 51% attack, and the miners' energy consumption may drop a bit. Bitcoin will just keep working.
What about the halving after that? Or the second after that? In less than 20 years from now, the block subsidy will be less than 0.1 bitcoin. Block space
has to be valuable.
If they achieve that, it would be akin to RFC 1918 (NAT) in IPv4 -> lots of computers sharing the same 32-bit address.
That's a good comparison. Solutions that group users in a single UXTO are in very early stages, and will probably never be implemented without softforks. You can check out
CoinPool, it's one of them. It can work more efficiently and privately than LN, at the cost of high interactivity.
Another cool L2 is this:
https://arkdev.info/docs/learn/intro. Its focus is fast, cheap and confidential transactions. Requires a softfork as well. There are more ideas, and it'll probably be just one L2 which will reach mass adoption, if any. Each one comes with its own tradeoffs, the market will decide which tradeoffs are the best to have.