The problem is, I've known people who could do that with pretty heavy rules--if they already knew them well.
I really don't mean to bust your chops about this; its just that people's definition of rules light can be an amazingly moving target. As an example, back in the day it was fascinating to see whether people thought RuneQuest I or AD&D1 were more rules light.
AD&D had (and all D&D flavors have) the huge issue of played weight vs RAW weight being a wider gap than almost all other RPGs, save maybe Traveller and Pathfinder. (Traveller due to intentional modularity, Pathfinder for having a fanbase of mostly people who think of it as D&D and treat it the same way.)
The greatest issue of light vs heavy is that so few people have seen the truly heavy games - games where every mechanical interaction is a unique rules case, where many interactions require table lookups on bespoke tables, where there are lots of special case rules that are unlikely to be used by 90% of tables, where detail levels result in dozens to hundreds of hit locations, where complex formulae encourage use of an HP41C or TI-60 (programmable sci-calcs) or a computer.
I don't see many new games that even really approach a true 4 of 5...
And, as for AD&D 1E? I'd put it, RAW, as about 4 of 5; RQ 1e/2e as at the border of 2 and 3... But AD&D1 game as played (GAP) ranged from 1 (the FKR type users) to 4 (rare, but plenty did exist), with some using all the whiz-bang articles in Dragon pushing their GAP up to a 5.
RQ 1e/2e, like AD&D 1e, have organizational issues, but not as many. Personally, I'd put it on the border border of weights 2 & 3, but the GAP being fairly consistently 2. It's real issue is not rules complexity, but that the game was focused upon a details-rich setting: Greg Stafford's Glorantha. RQ was intimidating due to the sheer scale of the setting...
RQ3, however... much better written, but also more complex. Definitely a 3.
But that ignores the added heft upon players of deep settings.