The decision making process you go through in producing an MEL is the same as you would in dealing with an individual problem day-to-day ie "is it sensible to get home with this issue and fix it later".
The idea behind generating your own is twofold:
First, it allows you to sit down and really review the aircraft's systems when you have your schematics and manuals there to decide if you're still maintaining a level of safety that you consider adequate. That probably isn't as big an issue on a day VFR carb-and-mag airplane... but on something with a dual-battery dual-alternator electrical system, electrically-dependent (e.g. EFI) engine, and IFR glass, for example, things could be a lot more complicated.
Are you ok flying without that standby alternator? What are your abnormal procedures for doing so? What happens if you then experience another failure? Are all your avionics tied together on a bus, or does each EFIS screen "manage" certain components? Those questions are a lot easier to answer with your wiring diagrams, electrical load budgets, and operating manuals on hand instead of trying to remember all that on the fly. I know on the big jets at work, there are non-MELable items which aren't obvious--you think "surely I don't need that, it's just XX widget!" until you really dig deep into the schematics or consider interrelated failures and realize that with XX failed, you have no redundancy in safety-critical systems.
I want to do my engineering when I have my engineering hat on, not when I have my pilot hat on. It's like the rule for not troubleshooting failures in flight--deal with the problem and troubleshoot on the ground, instead of playing mechanic when you're in the air.
The other idea, as I said, is to exercise some form of self-discipline. How many times have we read stories from pilots who let get-there-itis push them into flying with inoperative equipment, when they should have stayed on the ground? Yeah, ideally we'd all exercise superior judgment all the time and never let ourselves be pressured (or never pressure ourselves) into making a bad decision. But the draw of "oh, I'll be all right... I'm almost home and then I can fix it tomorrow" can be awful powerful. We regularly see large jet pilots and operators asking "we're on a trip and this critical widget broke; it's not in the MEL but can we finish our trip anyway?".
It might be of even greater benefit if you ever let someone else borrow or use your airplane, if it's a shared/club airplane, or if you ever sell it, especially (again) if you have complex or non-traditional systems.
You might know the airplane's systems intimately and be an experienced pilot with supreme judgment at all times... but the next guy might not be.
Of course, making such a list is only worthwhile if you're going to stick to it. Sure, it's unofficial and not legally-binding, and so you could just waiver yourself around it when it's inconvenient... but down that road lies "normalization of deviance", and that's somewhere I don't think we want to tread.