IMO the most interesting takeaway from this is the 2013 update:
that working in consonance with gender differences can help to boost achievement for both girls and boys, even in a coed classroom
Which means that we can't simply say that there are no gender differences - it doesn't work, probably because of the interplay of societal and cultural factors which perpetuate/reinforce existing biases
and also this part:
They randomly assigned 401 8th-graders either to single-gender physics class or to coed physics class, for one school year. At the end of the year, the girls who had been randomly assigned to the all-girls classroom were more engaged in physics and less likely to agree with statements such as "physics is for boys." Girls who had been randomly assigned to coed physics class were more likely to agree that "physics is for boys."
Which means that for certain limited set of subjects where gender bias is already present in the culture of the society, this methodology could potentially be used to help in removing that bias, (as a temporary transitional state)...