Ah, yes, the inevitable "I could do a better job" comment. Go on, then. Show us all how it's done. Except you never will.
You had quite a long post there and i can't respond to every single question but i think you understand the issues involved.
Satoshi did a better job. If they would have just left it that way. But instead they had to introduce things like OP_RETURN, segwit and taproot. What do you think the end result of all that is going to be? that's right. a huge mess.
I don't hold satoshi responsible for all this mess though. If people want to store data on the blockchain, they should have to pay through the nose but then they would never do it in the first place most likely. Just like it never caught on before all these upgrades took place. "upgrades".
Satoshi showed us how it's done.
You talk as if OP_RETURN, segwit and taproot bring no advantage. Should we remind you to few of these facts?
1. OP_RETURN created mainly to reduce P2PKH abuse by encoding 20-byte of arbitrary data as public key hash. It means less UTXO which will never be used.
2. SegWit practically solve transaction malleability and quadratic sighash problem.
3. Taproot let you only reveal part of the script. It means slightly better privacy and less TX size.
OK, so how is your plan?
Surprisingly, the plan would be similar to the original bitcoin. Only a few transaction types. No OP RETURN no nothing like that.
The challenge is that you can actually not allow anything which can be used for arbitrary data. For example, probably publicly visible fields for values with a large number of digits could be enough. Public key hashes too, like I demonstrated in the post I linked above.
Only have one transaction type. Pay to public key hash. That really is all I would have. The simpler the better and less exploitable it is. No upgrades for higher level functionality, nothing like that. Only bug fixes if necessary. But no new "features". Doing one thing and doing it well is all that really is necessary. Bitcoin has kind of lost that with all these ugrades and things. And we see the end result.
Thank you for the question.

Only P2PKH? I guess we should say goodbye to multi-signature address, address with "inheritance" feature, LN, sidechain and other innovations.