It is axiomatic that satellites in the same orbit travel at the same speed.
Your axiom is not axiomatic.
The idea that there is something called an "orbit" that a satellite can be "in" is a simplification, just like cars are in lanes. They are, roughly, most of the time, but in many countries they are all over the place, without discrete "lanes" evident nor consistent speeds. That's a much better model for thinking about satellite trajectories.
Once you start discussing a satellite's mass, then any mass at all is going to change the motion of everything else in the solar system a little. There is no minimum mass that starts to affect things.
See answers to Does launching a device into orbit change earth's orbit? for example.
As you increase the mass of the satellite, the orbit of the Earth and the mass around each other continuously changes, different paint chips move the Earth different amounts. Likewise the effect of the two paint chips on each other. There is no minimum. Certainly if one gets very massive it will start to noticeably perturb other objects in orbit, but that threshold is only related to how carefully you look and how little you can notice.
If so, where is the tipping point.
There is no spoon tipping point: YouTube
Your question really just asks about how long an approximation or simplification is valid, and that's purely up to you and how much error you can tolerate by using your approximation.
Also:
While you would like to imagine that there can be two satellites in the same orbit but one behind the other, the problem is that the Earth's gravitational field is not uniform, so it will not follow in the same path. Orbits are not perfectly closed or repeatable because of this.
It's a hard task to get out of the idea that there are really such things as fixed orbits, it's a jump of intuition, but that's the reality; there isn't. All you have is a gravity field with it's lumpy deviations, plus drag, solar pressure, and the Sun's and Moon's gravity (plus more), and a sophisticated numerical propagator that calculates step-by-step how a body might move in this complex and ever-changing field of accelerations.
Welcome to the edge of space!