I think the probability is high that Lightning won't be the general purpose instant payment tool it was described in the euphoria phase in 2017/18 when it launched.
The current stagnation could be explained by the low fees, but I don't think that is the reason, because it struggled to grow also in 2023 and 2024 when fees were high during the Ordinals craze. According to
BitcoinVisuals, Lightning channels nodes stats were even declining during that phase. It is possible that it was simply too expensive most of the time to "just try it out", and in addition some may have simply sold their LN balance when the price started to grow.
For now I have more hopes about Ark and sidechains that they could solve the scaling problem more efficiently.
But Lightning has also some advantages both compared to Ark (it is independent of a "provider", even if Ark providers are trustless) and sidechains (where you have additional risk due to an alternative consensus system). I think it has a future as a "prepaid card" type bundling option. For example, if you get your salary with Bitcoin, you could transfer a part of it each month to LN, and then use it for payments and so you only have to use the fee once. It is also useful to transfer BTC from exchange to exchange, but that is more a feature for heavy traders.
Of course all that depends on Bitcoin becoming more popular for payments again. The current relatively high number of transactions in the last months may be hinting at that (it seems to be not all Runes and similar stuff), but it needs to be confirmed.
Ideally we should have a wallet that could integrate every Bitcoin L2 solution:
- on-chain payment: for large transactions like getting your salary, or international txes.
- Lightning: as a prepaid card for small purchases.
- Ark: for people who only hold relatively small amounts but transact often, better alternative to CEX holding.
- Sidechains rollups tBTC-style decentralized bridges: similar to Ark; but also useful for people who need advanced DeFi capabilities as they often integrate EVM or similar advanced scripting smart contracting.
If this setup was used by most Bitcoiners then we could easily have 10-20 times more transaction capacity, and that should be enough for the next 10 years.