Does types being terms imply your dependend theory is considered polymorphic?

In the introduction of the book by B.Jacobs, "Categorical Logic and Type Theory" (it's online here), he classifies type systems into three general flavours: Simply typed ones, depended typed (term depended types) and polymorphic types (type depended types). He says also there are also mix types.

Now if you start out with a dependently types theory and introduce transitive universes, hence forcing types on the level of terms, are you automatically speaking of a polymorphic type system then?

These three “flavors” are usually presented as the lambda cube. They aren't really flavors but properties that can be combined independently, so there are not $3$ but $2^3 = 8$ flavors. In fact you could go as high as $4$ properties yielding $16$ flavors:
• terms depending on terms, i.e. functional languages (e.g. $f : x \mapsto x+2$);
• terms depending on types, i.e. polymorphism (e.g. $\mathrm{head} : [X] \rightarrow X$);
• types depending on types, i.e. type operators (e.g. $\mathrm{list} : X \mapsto [X]$);
• types depending on terms, i.e. dependent types (e.g. $\mathrm{array(int)} : n \mapsto \mathrm{int}^n$).