I still doubt that a stablecoin company can really compete with an established bank in their specialized fields. IMO it's much easier for a bank (once regulation is clear) to set up a new stablecoin themselves (as Societé Generale did with the EURCV) than a stablecoin company emerging from the crypto sector entering the loans/credit market and outcompete banks in this field. Banks have a lot of contacts to major companies in their countries, which are probably the most important thing in that business.

But maybe it's not that important who runs the stablecoin, if it's a "crypto firm" or a "traditional bank". I think we will definitely see more companies that will operate both - stablecoins and "business banking", with retail banking becoming less and less important.

The thing which makes me doubt most that Central Banks will disappear is however another thing: I guess that with a license which allows you to give out loans without a 100% reserve (i.e. fractional banking) is a big competitive advantage in the credit sector. And this will allow Central Banks to still control monetary policy more or less in the same way than now, only that their counterparties will manage also stablecoins.

From a "monetary" perspective, a stablecoin isn't much more than a traditional bank-based payment sytem, only with another technology, but with a similar relation to the underlying currency. The biggest difference is currently that a stablecoin can be held by people who aren't customers of any bank: you could buy USDT via P2P and store them in a non custodial wallet. And that can be indeed a competitive advantage too, but only if P2P stablecoin trade isn't banned.

Yes, such centralized apps encompassing several banks' accounts under an unified API already exist for some time, I've seen them both in Europe and South America. The only step left would be to integrate stablecoins in that offer. That's mainly a regulatory issue.