Design by Exception (or not)

Over a decade ago I posted about "exceoton based design" (here and here). Quick tl:dr... I didn' like it the and I still don't like it. I've thought about it plenty of times in the time since then, but the big games in the hobby (Dungeons and Dragons, and Pathfinder) still base themselves on it.

Oe of the key differences between minimaliust game designs and maximalist game designs seems to be the idea of wther a general game rule can be applied to everything, or whether you need specific rules to cover specific situations. There's definitely a continuum between the extremes... do you write a one page system like "Honey Heist" or "Lasers and Feelings", where a single mechanism of play defines all actions regardless of how well those actions might fit the mechanisms?...do you write a sequence of books each with a few hundred pages in them (There's a few of these, and entre business models based on the "supplement treadmill")?

I've tried writing multi-supplement games a few times in the past. The Law got three or four supplements written and laid-out, I think I only ever published one of them (with the Beneath the Glass and Steel zine sequence serving as unofficial supplements created as an "in world" setting guide). Shattered Heroes was developed with the assistance of my high school gaming club, and we had a half dozen supplements ready to launch with that game...but that had other issues. In each of these cases the core game was written in a small format, 16 or 32 pages, with similar sized extra supplements detailing an extra rule or two, and new ways to integrate those ideas into the core systems. 

Where am I going with all this?

The basic idea is that I like a game to have a limited number of coherent systems that will cover the vast majority situaations that arise in the storytelling, but ensure that those systems make sense. I've been toying with systems for both Walkabout and Familiar for a decade or so, the origins of the system were released in FUBAR, moving through various playetests and changes... but the fundamentals remain the same.  

Not one system that is badly showhorned into every situation...not a binary result fom a d20 plus modifier versus a difficulty...not a 2d6 with a good result on 10+, a mixed result on 7-9, or a bad result on a 6 or less. 

Not the other extreme with some systems using twenty-sided dice, others rolling percentiles, some rolling straght for a result, others on tables that look up other tables... and ccertainly not a situation where you roll on a table in one book and it refers you to a table in a ompletely different book (I'm looking at you Palladium).

I've also previously discussed the idea that the average person can handle 7 things in their mind at one time, and have mentioned how I design character sheets to account for this. So this all links together. 

A single sheet of character details. 

A single sheet of basic rule guidelines, explaining no more than three or four systems that cover all elements of play. 

A rulebook that explains how you can mix-and-match these systems to simulate all sorts of events that might occur. (Along with backgrround details for the setting, and plenty of story ideas)

If there are any other sourcebooks, they just provide new ways to combine the existing rules in different contexts to reflect events that may not have been covered in the original rules. I know I'm not going to gete everything into a single starter book, but I'm wondering whether my plan is still feasible.