I think you just need to run the align command with the appropriate parameters (to set a cutoff) and the syntax (to define residues for the alignment base). This tutorial gives a lot of details on these subjects: http://pldserver1.biochem.queensu.ca/~rlc/work/teaching/BCHM823/pymol/alignment/ and although pretty limited in scope, this wiki page might also come handy: http://www.pymolwiki.org/index.php/Align

For example in your case, all you need is to set the cutoff=1; e.g.:

PyMOL> align 1kao & i. 100-200, 1ctq & i. 100-200, cutoff=1

and you will see the RMSD score in the output printed within the console.

align command is not so good. There are a few other commands of this purpose: cealign, super, tmalign etc. http://www.pymolwiki.org/index.php/Tmalign

Align, Super, CEalign, etc, all these command take selections so you can use for example align protein_A and ss h, protein_B and ss h to align the helical regions of each proteins. Be aware that PyMOL will try to optimize the fit (by removing atoms) and reduce the RMSD value. You can tweak this using the cycles option (not sure if available in all alignment commands).

If you only want to calculate RMSD without making any structure alignment or fitting, you should use rms_cur function.