IMO and IMX porting core mechanics from system to system fails more than it succeeds. core mechanics are simply wovern into the foundations too much.
Now, on the general notion of breaking things down to a core mechanic of multiple actions etc, i think it has merit and also that in order to work best it must be done whole hog, not piece-meal half and half. By this i mean it must be comprehensive not just partially worked into the other stuff. No "you have three actions... oh and a reaction and a talking..."
To illustrate i will avoid the term "action" and go with:
The System of Many-Does
Each turn your character gets a number of "Does".
Each "Does" lets you say "my character does this..."
Examples of "does" can include:
My character does this (draw a sword"
My character does this (attack with my sword"
My character does this (cover myself from attacks of opportunity)
My character does this (keep an eye out for opening to strike if they withdraw") (aka take an attack of opportunity.
etc etc etc...
key being that this puts "going all out", "being cautious", "ready to..." (actions, reactions, bonus actions etc etc etc all into the same basic framework which lets you dial-in your character's degree of focus or divided focus turn after turn after turn.
this likely means more than three "does" per turn.
It could even go so far as to include "saves" as in spending a "does" gives you bonus on a save for that turn replacing the core "this guy gets better saves than that guy" locks.
if you set "does" to six, the a "spread" would mean you could move, attack, "bonus" act, be on wary (one reaction), and be on guard for two saves (one does each) letting you get a very traditional 5e turn.
But it would also allow you to do more diverse combos of things.
A higher number of "does" and dumping tons of "default assumptions" like "favored saves" (or using my shield for Ac boost or attack) can also make "casting a spell costs for each component" more viable...
Classes, races, feats or whatever could cover a lot of their features with "free "does" of limited type." maybe rogue gets a free "does" for "better dex save" but maybe halflings get a free does for "any one save type per turn".etc etc etc.
this could be a "complex to write out" but "easy in play" core mechanic that allowed a great deal of control over all-in to all-out and everywhere in between choices for the players.
It could be workable... has merits... a sort of simple to grsasp and maybe difficult to master.
but again, has to be at the core of all things.
**THAT ALL SAID** where i think it has a potential fail or epic success is how well the system integrates it for "non-combat" challenges and objectives.
One might try and define/present/resolve "non-combat" objectives and challenges as simply the same resolution on a larger time frame... letting your "does" count cover say an evening for "social challenges" and with a robust list of "social "does" actions/reactions, etc.
this, if well done, could effectively create a single "combat" mechanical system to resolve both combat and non-combat challenges. The "social" equivalent to "i wear heavy armor" (resource) might well be "i buy lots of drinks" and the equivalent to "finesse weapon" might be "an appropriate gift for their station."
Side note: I have liked in the past when systems had say nine attributes - three physical, three knowledge and three social... and in each had a stat for "power", a stats for "skill" and a stats for "toughness" that worked much like how STR, DEX and CON work for the physical challenges. They did make it obvious and mechanically beneficial to cover multiple attributes in a cluster and did not as easily make any "type" of challenge a "one stat" affair.