First off, I don't have the background to calculate actual values on how much harder other landing sites are to reach, but I'm happy to add my thoughts about it.
There are a few different mission profile options to consider (adapting names from Apollo mission modes):
Your entire ship lands, leaving nothing in orbit, then on return all or part of your ship leaves the moon to return directly to Earth (or some other destination), again leaving nothing in orbit.
In this mode your lunar orbits don't particularly matter: because you're not leaving anything in orbit the landing and ascent orbits don't need to align. You just need an insertion orbit that goes over the site (easy to set up while approaching the moon without requiring plane changes when you get there), then on departure you just need to take off in whatever direction is most efficient for where you want to go (and again do course corrections while in transit instead of plane changes while in orbit).
Note that this assumes you have no issues that could delay landing such that your orbit and landing site no longer line up (either making landing at the planned site impossible or costing more fuel to correct, and how much fuel buffer did you have to include specifically for this reason that you could have left out given a polar/equatorial site?). Having the ability to make fallback plans that still result in a successful mission is generally a plus.
To save the fuel cost of landing and later launching your return stage you leave it in orbit, taking only a lighter-weight lander to the surface.
Given that you eventually need to rendezvous with your return stage you have a few options (drawing from what you suggested):
- Plan the duration of your stay on the surface arrange your orbit to align with the landing site for both descent and ascent. You don't have to wait for a full revolution, you just have to wait for your landing site to be in the plane of your orbiter (though you do save some fuel if you can take advantage of the rotation of the moon by taking off in a prograde direction which would require waiting the full ~29.5 days). I can best explain this with a picture (pardon my lack of artistic abilities):
You're not landing at the pole so you have orbital inclinations anywhere between your latitude (aligns once every 29.5 days) and polar orbit (aligns once every 14.75 days alternating ascending and descending passes). The closer your orbital inclination is to your landing latitude the closer the time on the surface will be to 0 or 29.5 days (land on an ascending pass for a descending pass to be overhead soon, or land on a descending pass and wait most of a month for the ascending pass to come around). For instance, if you moved the landing site to the other part of the blue line that crosses that latitude it would take 20 days for the next orbit that aligns with the landing site instead of the 10 days that I represented in the picture.
Also, here's a picture of orbit traces from Apollo 15 illustrating how much the moon shifted underneath the orbit of the command module in just 14 orbits.
You're saving fuel by not needing any plane changes, but your schedule is tightly constrained by the orbit of your orbiting stage.
Have a high apoapsis so plane changes are relatively cheap. This means you'll have more velocity to burn off on descent and more velocity to make up on ascent. I suspect the overall result would be cheaper in terms of fuel than plane changes to align with a circular orbit, but I don't know the math to check this offhand and it's still not going to be cheap.
Plan your landing site and orbit such that lunar gravitational anomalies take care of the plane changes of your orbital stage for you. I don't know how good a map of the lunar gravitational field we have but I think you'd have to be crazy to plan on this.
Landing on the moon is all about tradeoffs:
You can land your entire ship to gain convenience (not needing a lunar rendezvous) and maybe eliminate complexity to reduce risk, but it'll cost you more fuel to land the additional mass.
You can leave part of your ship in orbit (reduce lander mass) to save fuel on landing, but now you have to align orbits and rendezvous with your orbiter when you leave so you have to consider the latitude of your landing site.
I'd say a less accessible landing site could certainly be worked with, but you have to decide if that landing site adds enough value over a site that's more convenient to access for it to be worth the extra trouble.