If you just scale it in your slicer you might have to go with a smaller nozzle, yeah. But if you change the source file (in FreeCAD) you can probably change the overall dimensions but leave the depth and gap the same. That would let you print smaller dimensions (length/width) with a 0.4 mm nozzle even. You'd also have better control over the screw size. If you did want a smaller gap, it would probably also require thinner layers (maybe 0.08 mm) to make sure the pivoting layers are sufficiently offset. Print slow to give each layer time to cool.

Hmm, well, it probably depends what you mean by "actual door." I designed this for an RC tractor I'm building, so it does actually work, but it's supporting a very small load (a ~1/6th scale door printed in PLA). Picture attached of it "on the job." :-) If you mean a hollow-core interior door or anything more substantial (solid-core, metal, exterior, gate, etc.), I mean, I probably wouldn't, personally, even if it scales correctly to that size when printing (I haven't tried it). When you connect a door to a hinge you have two big forces being applied to the hinge: an upward force counteracting the weight of the door due to gravity and a lateral force due to the center of gravity of the door itself. While I would expect an upright-printed hinge to be stronger than a side-printed hinge in both of those circumstances, it's still layers of plastic at the end of the day. It won't be anywhere near as strong as metal or even molded plastic. Also, from a convenience perspective, while you can normally (temporarily) remove a door by taking out the hinge's pin, since this hinge has no pin, you'd have to physically unscrew the door if you needed to do that. If you do try it, make sure you increase the padding around the screws and depth substantially to try to spread the force out. As a reminder, this design comes with no warranty! ;-)