How do they fit the Randle Sevick's Equivalent circuit for multiple data (Nyquist Plot). Since most of the journals shows a single equivalent circuit representing for multiple nyquist plot .
I believe that you are confusing the Randles-Sevcik equation with the Randles (equivalent) circuit.
In journals, for a given experimental system, we expect the same phenomena to take place, so a single equivalent circuit is used. EIS data for different systems can be fitted as long as these same phenomena take place (e.g. the EIS of 0.1M, 0.05M and 0.01M HCl would show the same phenomena but would differ in terms of the parameters of the components of the equivalent circuit).
If the circuit has several capacitors, each should (should as many authors tend to use too many components in their equivalent circuits) represent a different capacitive phenomenon, so you only need to identify, which one stands for Cdl.
From my understanding, I beleive that it answers your question.
Multiple curves can be represented by same Equivalent Circuit as long as it belongs to the same system and same processes are happening in the system. For eg. if you make impedance spectrum for different temperatures of the same system. Then you can fit those family of curves using same Equivaent Circuits with an appropriate initial condition (Batch Fitting)
The simple(st) equivalent circuit (EC) models, having less time constants, are preferred to avoid redundancy. All curve-fitting software should report some measure. A measure is the X2 (chi-squared parameter) benchmark. The value of X2 should decrease[1] (say by tenfold, as an indicative criterium), when a new circuit element is introduced into the circuit model, creating a lower X2, but a more complicated EC.