I personally have never used these, but I've always been curious to how accurate they are. Like you said they should be relatively easy to control. Granted, the stepper based program is already up and running, so you would no doubt spend more time coding if you go this route, but probably less time building the machine because the actuators are finished and ready to mount.
It would be nice if there was a datasheet somewhere that gave some info on this type of actuator's accuracy. I've also wondered if their positioning varies some depending on both the temperature and on wear on the potentiometer over time.
How do the potentiometers even work in these. Do they just give you the measured resistance so that you can count the revolutions or what? If that's the case then they aren't really saving any work. I think there are types of linear actuators that do give you the length when you measure the resistance, but it's not obvious to me that these do. Like I said though, I have no experience with these things.
A greenhouse is generally a pretty big target, so accuracy might not be a huge deal in this case anyway, especially if the heliostat isn't set up very far from it.
The biggest disadvantage with linear actuators in general is that they restrict the range of motion for the machine. Whether or not this really matters depends on both your latitude and the time of the year you are most interested in getting energy from the sun. Where the target is in relation to the heliostat can make a difference too.