Sure, but let's not pretend many people are actually using Bitcoin for on-chain payments. I would much prefer that people were, but it's not the case. In fact, many people on this very site have told me that they want the network to be more expensive for people to transact on. Or they say to "go use lightning" if you want cheaper transactions.

Am I wrong or somehow misunderstanding how Bitcoin works, in thinking that even if ordinals were to go away completely, Bitcoin would still be dealing with these same problems of congestion and expensive transactions when met with an increase of people transacting on-chain?

Seems like this has brought to the forefront of people's attention, the issues with scaling L1 bitcoin or providing a better alternative to lightning for onboarding the people we expect to be using Bitcoin for a true future economy. I think a lot of the effort that's been spent attacking ordinals and the people engaging in it (especially when you can encode data into the public keys themselves on-chain, encoded as multi-sig outputs - https://github.com/mikeinspace/stamps/blob/main/BitcoinStamps.md - which would take up even more space than Ordinals, btw) would be better spent focusing on building viable scaling solutions for bitcoin so that this isn't even an issue anymore.

Some of the people building out alternatives to lightning network are coming up with very interesting stuff, like Chainway who are creating a true trustless, programmable scaling layer by way of a ZK rollup - https://medium.com/@chainway_xyz/a-sovereign-zk-rollup-on-bitcoin-full-bitcoin-security-without-a-soft-fork-ca0389a0b658 - they are just one several such projects to keep an eye on, IMHO.