Low noise transistor s2p parameters vs cascade vs cascode

Wouldn't a true LNA have at least 2-stages to increase the gain before being modulated in a superheterodyne receiver? I am trying to design on ADS a low noise amplifier but with 2 more stages. First of all, wouldn't you also put a 2nd stage transistor to increase the gain in an LNA, or how would you do it? maybe you would just look for a single transistor that can give you enough gain with a very low noise? I have seen LNA cascode designs before also. Anyway my main point is If I put two common emitters/drain of the same transistors in cascade using s2p file (lets say I configured my gummel_poon model to replicate the s2p file at the specific frequency instead and used this). Would this work correctly if created a 2nd stage in cascade? also would the gummel poon model work correctly if put in cascode, since one is common base/gate, and the other is common emitter/drain would those two configurations throw off the s2p parameters and wouldn't be accurate? or would it be accurate?