Hi Moza,
You have most of it, but you need to add a step and change a sign error. The section terminated by R and C, I called that section A of length la, and the shorted section is section S of length ls.
At the point where the two lines connect, the impedance of section A is not ZL, which is at the right end of the line, but rather the impedance seen at left end of the line, which is
Za = Z0*(ZL + i*Z0*tan(beta*la))./(Z0 + i*ZL.*tan(beta*la)); (1)
and your shorted stub impedance has a sign error and should be
Zs = (i*Z0*tan(beta*ls)); (2)
I defined (same as what you have)
lambda0 = c/2e9; so
case 1 la = .110*lambda0; ls = .095*lambda0;
case 2 la = .260*lambda0; ls = .405*lambda0;
If you add the new Za and Zs in parallel and keep going you will reproduce the plots.
Incidentally, with vector f a for loop is unnecessary if enough .* and ./ are used, which it seems like you already have quite a few of but not all.