I might be misinterpreting him, but here is my take on it:
***'Since there is absolutely no way to uncover what processes take place within that pipeline, there is no way to justify any symmetry in the things I think exist to be a valid characteristic of reality itself'
'Thus it is that the first step in my analysis is to assure that my bare model displays every symmetry that I can think of.'
'By this mechanism I assure myself that my model will not attribute to reality things which could have been generated by the pipeline. '***
In other words, Dick is making an epistemological distinction and not ontological. He is saying, in effect: "I have no knowledge about the actual reality behind any symmetry I use (i.e., ontological distinction is negative), but since I cannot assume any symmetry exists, I must work off only my knowledge of all the symmetries that I think that there are to me from my perspective right here and now (i.e., epistemological distinction). This epistemological distinction is just to work within my own limitations, therefore I am justified in setting up these symmetries based on those limitations. By this epistemological approach I assure myself that my model will not attribute to reality things which could have been generated using an ontological approach (e.g., symmetries actually exist in reality. By doing this, I can take advantage that the symmetries generated by my model without making any undue assumptions about reality per se".
This approach is all fine and good, but it again overlooks the conditions that he himself places on himself by taking an epistemological approach in one section of his paper, and yet still crosses into making ontological distinctions (e.g., reality can be defined). This is why Dick's metaphysical approach is way too inconsistent to be of any use. Frieden is much more consistent in this way.