Is it possible/advisable to have multiple state machines in an app?

For example, have a state machine for the entire app (i.e - startup as used in the sea of arrows) and then have another state machine that is used for some forms that appear in a popup window (i.e a sort of like the FSM as described in the puremvc.tv).

Or should I just use the state machine for startup related things and then have a separate mediator handle the form and it's various states?

copy and pasting over and over to make a bloated class is definitely not ideal.

that's the scenario i'm trying to avoid - writing more code that could ideally be avoided by just extending another class.

i have no desire to have one big proxy - what i would like is to have one proxy subclass another proxy but send out different notifications with different payloads/bodies (i.e. Cat vs. StripedCat).

I would like to extend a Proxy that is calling a webservice; however, I'm not sure if this is advisable.  The scenario is I have 2 webservices that expose the same webmethods; however, they have different return types, i.e. the first service might have methods that return a type "Cat" and the other service have methods that return a type "Striped Cat". 

i.e.
catservice.asmx
getCat():Cat
getOldCats():Cat

stripedCatService.asmx
getCat():StripedCat
getOldCats():StripedCat

Would I have to recreate the exact same proxy twice?

i would also like to clarify - by view i mean 1 component. (i.e a tree, tilelist, etc)

I am so far just reading and learning about puremvc and just wrote my first few lines in converting an app (very old AS 2 v2.0 components) to flex/puremvc.

so question 1:
is it best practice to have a 1:1 relationship btwn a mediator and a view?

question 2:
if i have a webservice with 3 methods, I should also have 3 Command classes?  i.e. webmethods to Commands are also 1:1?