I don't want to jump in with too much technical jargon, but I deal with this sort of geometry for a living, so I have a tendency to assume people know what I'm talking about. So, get ready for the world's longest answer...
I wouldn't bother with easements, well, not in the grand sense that most people think of easements. The way model railroaders have been taught to think of easements (or spiral curves) is really how they're used in conjunction with superelevated curves, but the superelevation part usually gets left out of the discussion. Most people tend to think of easements in these large, slowly sweeping transition terms, when the truth is all real-world curves have easements to one degree or another.
Just to clarify what I'm talking about, the spiral curve is used to as a "bridge" or a transition between a straight (or "tangent") section of track (or highway) and a radial (or "circular" or "fixed radius") curve section. The length of this parabolic curve is determined by the circular radius and the design speed of the curve. The design speed and radius defines how much the outer rail (or outer lane of a highway) is elevated or banked above the inner rail. Throughout the circular curve section, the difference in elevation (superelevation) between inner and outer rails is constant. On the tangent, both before and after the curve, the inner and outer rails share a common elevation. It is throughout the spiral curve section that both the trasition from flat to superelevation and the transition from straight to curved occurs.
Now, back to the question at hand. If you think of a track gang setting down a rail at the end of a section of tangent (straight) track, fishplating and spiking the end in place, then having to wrestle the rail to bend it into place while it's being spiked as the curve begins, you might imagine the little difference between joining a straight piece of sectional track and a piece of circular radius sectional track. In that 39 ft. length of rail is the complete transition from straight to curve, which is unlike the two pieces of sectional track, where one piece is straight as an arrow and the other piece has a common radius point that can be measured from any point on the rail. The prototype's rails have to slowly ease into that curve because they are bent into shape - not by machine, but by lateral force. The transition from straight to curve might only occur in that one 39 ft. section of rail, or it could be stretched out over a couple, but it's not abrupt like on model railroads built with sectional track. And this transition curve is defined by the tendency of the steel to stay straight, not by a complex curve formula.
Another thing I can tell you is that in my experience, narrow gauge railroads didn't design spiral curves into their alignments. Many of the roads and even some of the highways in southwestern Colorado are actually old RGS or D&RG railroad grades, so when you do boundary retracements out there, the survey data defining these rights-of-way is the old railroad alignment. All the ones I ever dealt with, including the tight curves - were all radial curves. Now, you can ride on some of these railroads and feel the smooth transition from straight to curve while in the back of your mind you know that "this is the point of curvature of a 5 Degree curve to the right." And I think this smooth transition is where the reality of bending steel meets the academics of tangents and radial curves.
Not only that, these railroads were in the business of getting the rails laid cheap and quick. The surveyors and track gang needed to be able to glance at the plans, turn the angle and shout to his head chainman "drive it!" and be done with it, not sitting on the dirt doing long division.
So, assuming you're not using sectional track (does anybody even make it in HOn3?), you can just go with the largest radius possible. But, when you're laying your track, try to imagine how the track gang laying it would have done it, the resistance the steel would have put against their backs, and use your eye to judge how smooth that transition between straight and curved can be. In the field, even a well-equipped surveyor with a crew of two, a decent transit and a steel chain couldn't have told the track gang anything more than whether they were "on line" at 100 feet or not. Once the points of curvature were staked, it was the steel that defined the curve, not the surveyor there on the ground and not the engineer back in the office.