In Bowers (2016), I argued that there are (a) practical problems with educational neuroscience (EN) that explain why there are no examples of EN improving teaching and (b) principled problems with the logic motivating EN that explain why it is likely that there never will be. In the following article, I consider the main responses raised by both Gabrieli (2016) and Howard-Jones et al. (2016) and find them all unconvincing. Following this exchange, there are still no examples of EN providing new insights to teaching in the classroom, there are still no examples of EN providing new insights to remedial instructions for individuals, and, as I detail in this article, there is no evidence that EN is useful for the diagnosis of learning difficulties. The authors have also failed to address the reasons why EN is unlikely to benefit educational outcomes in the future. Psychology, by contrast, can (and has) made important discoveries that can (and should) be used to improve teaching and diagnostic tests for learning difficulties. This is not a debate about whether science is relevant to education, rather it is about what sort of science is relevant.