As I've been thinking about what antennas will work well for me, one idea I like is installing a SteppIR SmallIR vertical at the rear of my property, and using a set of tuned radials for 20, 15, and 10 meters.
One reason this appeals to me is that it's easy to model (I've already built the EZNEC model using two radials per band; I'll post the charts here in a day or two). On the other hand, I have no idea how to model the other antennas I've been looking at, like the Cushcraft R8, due to lack of information about the antenna's construction, and my lack of modeling knowledge.
Although this antenna will be a few dB down on a dipole up at least 1/2 wavelength, it should do better to the east and west than my G5RVjr, which has major lobes to the N-S on the higher bands.
Additionally, the visual profile of the SmallIR is much cleaner than the other antennas; it's just an 18-foot high vertical pole, although the control box will be visible from the street if mounted high enough to avoid serious ground loss.
In thinking about this, I realized it would be very easy to construct a prototype of this antenna using some surplus aluminum tent poles I have laying around. So last night I did that. Happily, the aluminum pole's outer diameter exactly matches the inner diameter of some PVC pipe I have, so it was very easy to make the insulating joint. These tent poles have a spring-loaded button that pops into holes in the sleeve section to adjust the length, so adjusting for resonance doesn't involve any cutting at all.
Just in some preliminary testing last night with it leaning against a fence, the antenna has a nice resonance point at 14.1 MHz, with low reactance. This weekend, I'll temporarily mount the antenna close to the planned final location, and do some on-air tests when the band is open. I'll be able to A-B the antenna vs. the G5RVjr.