Finish the cabinet, then build a better one. You can’t skip to the second one.

The second attempt is better because of everything the first one taught you, and that knowledge only exists if you took the first one all the way to the end. Abandoning it near the finish skips the part where you learn how things actually fit, where mistakes compound, and what you would do differently. Experience cannot be inherited from a version you did not complete.

Take the project you have stalled on and finish it in whatever rough state it reaches. Write down three things you would do differently. Then start the second one immediately, while all of that is still in your hands.