jem-mosig comments on More Recent Progress in the Theory of Neural Networks