It would depend on what your actions are doing. In some cases maybe yes but in other cases no.
I have a set of actions where there are 42 actions because of different paramaters needed to adjust based on image size, direction, and density. Instead of 42 action, I could reduce this to 3 actions and then have the UXP coding handle direction and density based on user input. The sizing wouldn’t be needed because the UXP could get the image size and automatically adjust for things. So in this case there would be 3 actions that get played along with 3 places in between the actions that do a few steps to compensate for the parameters. The end result is a much easier user experience and more control too for more precise density other than just 3 settings.
So in this case, while the user would have 3 actions, they wouldn’t be able to put it all together anyway because they would be missing the important pieces of the puzzle which are the adjustment steps done by the UXP code. Sure, some people may figure out what is being done in the actions and be able to piece there own together. But that isn’t something I’m worried about for a bunch of actions I have that I want to use this method for.
I’m not saying this is the best solution for everything. I also have 10 plugins that won’t have any actions at all and are much more proprietary. However, the action method is an easy solution for things like this that are less proprietary and where the ROI to build it all in coding isn’t there to justify the workload doing it.