Can Motor Recovery in Stroke Be Improved by Non-invasive Brain Stimulation?

At the present time, there is enormous interest in methods of non-invasive brain stimulation. These interact with ongoing neural activity, mainly in cerebral cortex, and have measureable effects on behaviours in healthy people. More intriguingly, they appear to have effects on synaptic plasticity that persist even after stimulation has ceased. This has led, as might be expected, to the proposal that brain stimulation methods might be therapeutically useful in rehabilitation. The rationale is that physical therapy involves learning new patterns of activity to compensate for those lost to the stroke. Enhanced "plasticity" produced by brain stimulation might increase the ability to learn and enhance therapy. However, if things really were as simple as this, brain stimulation would be on its way to becoming a standard addition to treatment in all departments of rehabilitation. The fact that this has not happened means that something is not quite correct. Is the theory untenable, or are the methods of stimulation suboptimal?