tensegrity - a conflict of models

It's been mentioned to me many times that the human body (in the biomechanical sense) can be considered using a tensegrity model (I use "a" rather than "the", because it's possible to model the body using more than one specific tensegrity model).

However, there are some features of tensegrity models which are somewhat surprising, which seem to run contrary to the ways in which we conventionally describe the human body, and I was wondering if anyone had any comments on the following:

0. Is the body inherently a tensegrity structure (at all levels of scale), or is there a requirement for some sort of conscious decision to use the body as a tensegrity structure?

1. Tensegrity structures are composed of (purely) compressive members arranged in a discontinuous skeleton, joined by (pre-stressed, purely) tensile members. In other words, the compressive members do not touch, and the structures connecting them are in tension. If this is the case, then what we would expect to find is that we should be able to set up the body so that all joints are in tension (i.e. distracted, to some degree), rather than compressed! Do you think this is possible?

2. Tensegrity strucutres are torque free, so essentially, it should be possible to centre a rotation on any point in space (it need not even be on the structure itself!). Does this mean that we need to refine our descriptions of body movements (i.e. flexion, extension, lateral and medial rotation etc)? If not, the term "lateral rotation of the arm", would refer to an infinite class of movements, each with differing consequences through the fascial network of the body, and as such might not actually convey enough information to make it a useful description of what is actually occurring!

Any thoughts?