It is impossible to generate two different sets of natural frequencies for one analysis. You are analyzing the system as a whole. The eigenvalues of the system will be proportional to the natural frequencies, resulting in the list of natural frequencies you see in the second figure.

It would be useful if you provided a picture of your sketch and more information about the boundary conditions.

Did you assign density to both parts or did you add a mass element? I guess you may have just modeled the two parts with out density and added a mass element. The maximum number of modes you can extract depends on the nodes having mass not the total number of nodes. Further, it is equal to the number of independent degrees of freedom having mass associated with them.

Like Mir Aamir Abbas mentioned, the number of modes you can extract will depend upon the number of degrees of freedom in your model. Assuming that you have meshed the two sections of cantilever beam (therefore having a large quantity of DOFs in your model), you should also check the frequency bandwidth that you are solving. Perhaps only one mode is located in the frequency bandwidth you have defined.

Attached is the sketches as well as the results. I managed to make it into separate parts with 2 different sketches in the same XZ plane and extrude it with the thicker part frozen. However, I was expecting 2 set of frequencies (since there are 2 parts) but I can only get 1 set of frequencies.

Both parts are assigned with Young's Modulus and density.

It is impossible to generate two different sets of natural frequencies for one analysis. You are analyzing the system as a whole. The eigenvalues of the system will be proportional to the natural frequencies, resulting in the list of natural frequencies you see in the second figure.

Thank you very much for the feedback. Appreciate it