Quality code is it's own reward. Yes, clean code is more fun to write and work with. But, just think of it as a different kind of optimization problem. If the investment is not going to pay off in the future, then it's not worth doing in the first place.
PS: This is not giving up, it's about getting even better by becoming ever more flexible.
whoever works with you will have to bear the burden that you left behind because you think its a low ROI for YOU
I've known of or read the situation where someone opens some old code, thinks "Who wrote this hot garbage?", and then realizes it was them.
Did it work? Can you read and understand it? Then it has value.
The 3rd time you touch a function is often a great time to make it awesome. However, a lot of code never reaches production and maintainability is more of an issue for code that survives.
Finding the right line in the sand for your current job is an art. There is ALWAYS more you can do, and being smartly-selfish is inevitable.