There is no out of the box satellite trajectory models in Sensor Fusion and Tracking (as of R2019b).
However you can get to it with a little bit of work.
The idea would be to leverage the kinematicTrajectory system object. You can inherit from this class and override the method stepImpl. If you are not familiar with system objects, I recomment taking a look at this documentation page.
I am copying a code snippet of what this could look like to use keplerian trajectories:
classdef myTrajectory < kinematicTrajectory
Mu = 398600.4405;
methods(Access = protected)
dt = 1/obj.SampleRate;
state = [obj.Position obj.Velocity]';
k1 = keplerStateTransition(obj, state);
k2 = keplerStateTransition(obj, state + dt*k1/2);
k3 = keplerStateTransition(obj, state + dt*k2/2);
k4 = keplerStateTransition(obj, state + dt*k3);
state = state + dt*(k1+2*k2+2*k3+k4)/6;
obj.Position = state(1:3)';
obj.Velocity = state(4:6)';
function dstate = keplerStateTransition(obj,state)
vx = state(4);
vy = state(5);
vz = state(6);
mu = obj.Mu;
r = norm([x y z]);
g = mu/r^2;
dstate = [vx;vy;vz;-g*x/r;-g*y/r;-g*z/r];
In this code, I am using some properties from the parent kinematicTrajectory class: Position, Velocity and SampleRate.
You would then use this class by assigning it to platforms in a tracking scenario:
scene = trackingScenario;
spaceObject = platform(scene);
spaceObject.Trajectory = myTrajectory('Position',[8000, 0, 0], 'Velocity',[0, 7, 1.2])
This is just a starting point, of course you can try and adapt it to your needs.