BRC-20 indeed has a lot of optimization potential. It's so ineffective that its data footprint could be basically halved. But the problem is that there are practically no BRC-20 devs interested in optimizing it, they seem to simply squeeze out all money they can from uneducated buyers.
If you mean the "Bitcoin devs", it's not so easy. Segwit already provided some optimizations. Rollups, what I mentioned in my previous post, are a way to optimize their footprint even more, but as far as I know there is no way to implement them without code changes which are controversial in the Bitcoin developer community (covenants like in BIP 118/119, as far as I'm aware).
Actually the only way I know which is already available to restrict data storage on BTC is to completely change the protocol using a very restrictive privacy coin mechanisms like Grin does. But that would also cancel Lightning, atomic swaps and a lot of other useful stuff, and need a hard fork.
There is a long-term possibility mentioned by Gregory Maxwell in another thread: the storage of blockchain data could be done in a way that the "raw" transaction data isn't needed anymore for old transactions, but replaced by other kind of proofs (often with Zero-Knowledge proof tech). This would mean that full nodes would not have to store all inscriptions anymore, and would really be an interesting solution because it would prevent also attacks based on illegal data (military information, illegal pornography/violence and abuse imagery, etc.) stored in ordinals. I think the devs are open for this possibility but currently these techniques are in its infancy.
I've started a thread about those concepts some time ago and will update it as soon as I have news about new developments in this area. One of the most advanced projects seems to be ZeroSync at this moment.