Are you sure you haven't buried yourself in a technical hole of despair?
If you have, it may be faster to treat your existing version as a prototype and rewrite a highly simplified version (an "MVP"). You can probably reuse a lot of the existing code and fix all of the big mistakes that are dragging you down, like not using React.
Make it your goal to get it released by a certain date and stick to it. Ruthlessly cut features until it's possible to meet that date. Reevaluate every sacred cow. Reason from first principles about what your product absolutely must do.
In my opinion, if you don't know if you have a product-market fit, there's no point in effectively doubling your "bet" in the existence of the market by rewriting the prototype.
So you're not using the best possible tech, but neither scalability nor performance matters when there's absolutely no users. Once you get at least some traction you can make more informed decisions about the rewrite, but right now you'd be just throwing more coding hours into creating something that might not have a market.
I agree with the last chapter 100% though, just throw the baby in the pool and see if it can swim. Our launch version had a lot of features implemented as "one of the founders is going to do it manually if you click on the button". Then we automated the tasks we actually had to do more than once per hour.
To this I would add: show the code to someone else. It sounds like you might be the only one working on this project, and that in itself can be a very dangerous thing (especially just getting started out). Hire a contractor if need be, even if you just book a week of their time so they can sit and look over everything with you.
If you have, it may be faster to treat your existing version as a prototype and rewrite a highly simplified version (an "MVP"). You can probably reuse a lot of the existing code and fix all of the big mistakes that are dragging you down, like not using React.
Make it your goal to get it released by a certain date and stick to it. Ruthlessly cut features until it's possible to meet that date. Reevaluate every sacred cow. Reason from first principles about what your product absolutely must do.