While I agree generally with Bunny83, I think I can guess at your meaning.
Our confusion comes from the fact we’ve studied this type software extensively (I say type because I’m recent to Unity, but I’ve worked on various game and physics engines), and we know that collision itself is the first part of the physics engine, so it is physics. I believe what you’re saying is that you’re not moving objects by physics. The standard step which follows a collision is to calculate the forces exerted as a result of the collision and apply those to the objects such that they correctly respond to having collided, with reference to the nature of the materials involved.
From what I can see, you have the basic idea, but there are lots of tributaries in thought and design which are beyond the scope of a post. That can become the domain of writing physics engines, something that is barely within undergraduate studies of mathematics, physics and programming.
When objects collide there’s a counterforce that results, and if the direction of that force isn’t correct the results look odd if the intent is to present a realistic physical behavior. In games, however, this is occasionally unimportant. Your list of steps would exhibit this error when compared to real physical collisions, so you have to decide if that matters.
The subtle error you appear to have in your list, however, is the recorded position at collision. This depends on whether or not your collision objects is larger than your visible object, and the speed is such that the results happen to look right. Instead of recording the position at collision, one must consider either to use the previous position, before the collision, or to calculate that position which results in contact minus a very minor margin, so it looks as though the objects touch but don’t overlap.
If the objects are simple enough, you may do well enough with a capsule or sphere collider. If that can do, you can avoid physics collisions altogether and work merely on the distance between the center of the objects in question relative to the radii of the spheres and/or capsules. It is much faster, but can be an oversimplification.