This was originally intended as a comment on Patrick Stuart's blog post WHAT IS FLOW CONTROL?? (AND WHY SHOULD I CARE?) but my comment was too long so I'm just making it it's own post in response. You should definitely read that post first or this response won't make sense.
I appreciate the self awareness you demonstrate in questioning the extent to which you actually prefer a greater sense of freedom even if it's made a lesser experience, and likewise differentiating between actual freedom and the sense of freedom. These are both critical, I think, to any kind of discussion on this topic.
While I dislike the extreme linearity or narrative constraints of Paizo/modern WotC design and a lot of PbtA/FitD/storygames respectively, it's also an easy straw man/false binary that I can already kind of sense people will fall into, so while I don't think that's what you're doing, I just want to get that statement out of the way.
I'm inclined to think that flow control is underrated in the OSR scene, in part as an overreaction/overcorrection to the bad and/or very extreme examples above (I don't mean to say they are as a rule bad, but when leaning into an extreme it's easier to go bad).
In this particular example with the house, in the absence of more specifics, I might be inclined to agree with you to some extent that it's better suited to an open-ended design. That said, a few key flow control gates could have a profound effect, especially in what is otherwise an open-ended design, and so if used conscientiously, both of these tools can and should be utilized together.
People tend to conflate this idea of flow control with linearity, but I actually think that's a mistake. When done poorly this is true, but when done well, it can actually create more possibilities, or rather, even to the extent that it is constraining, it provides a sense of greater possibility by making the possibilities more salient (getting back to that idea of actual freedom vs. a sense of freedom).
Frankly, what I've often found with OSR design, is that it is so open-ended, that most people aren't clever enough (or at least not clever enough on the spot) to actually think through the possibilities and do something interesting with them, so they all kind of play out the same way, not unlike sub-par RNG Dungeons in some videogames. That actually probably comes off more insulting and provocative than I intend, it is genuinely difficult for to work without constraints, and some people perform better under those conditions than others, and I don't mean to suggest that they aren't at all clever, but anyway...
By having some carefully considered flow control gates, where in the first case the Designer, and then the GM, and then finally the Players, know that there is some kind of limits in place (or have some sense of them), more possibilities unfold or become salient. I've heard it said in the context of videogames, sports, and children's games, but I think also applies to TTRPGs, that games aren't defined by what you can do, but by what you can't.
If the designer knows that no matter what X has to happen before Y, they can design more interesting leads for how X gets to Y, or things to encounter between X and Y, or in how Y branches out into Z or A (but absolutely not B! But that's ok, because X precludes B; in other words, the branches on Y are a function of X). It goes much deeper than that, I would argue this same principle of flow control vs. open-endedness in scenario design also applies to mechanics design and many other facets of a game, and not just games but also literally systems in the abstract, but that's also all an aside...
But back on the topic of flow control in scenario design, the GM can better prepare for the kinds of actions the Players may make, and can more easily improvise if Players do all the whacky stuff that Players tend to do. It might be that the Players do something which invalidates X, but if the GM knows that it all needs to lead to Y, they can improvise towards those ends, and do so more easily than if there were no framework or no constraints whatsoever (circumstantially, I don't necessarily mean this as an absolute).
And the Players too, if they're given these various leads by the Designer or GM, they can still choose any number of ways to approach the situation, or choose to come up with their own approach, but at least they have some leads to go off of, because there is some end in mind.
And nothing says, despite all this, that Players can't totally upend the whole thing, because that's just TTRPG, but having some degree of flow control and intentionality with that is only mutually exclusive with freedom if Players aren't clever enough to find their own way or GMs aren't clever enough to deviate, and good Design should lower the bar for both regardless.
Anyway, I realize some people get really heated with this stuff, so I welcome thoughtful discussion to this response not unlike Patrick's post in the first place, but please I'm not looking for throat cutting argumentation here.