Comment on neuromorphic completeness