I have not looked at this particular study project very closely, but Caltech has some smart researchers and I am sure they have an idea worth investigating. I did my senior design project on a clean sheet design using blown wings, and the idea works pretty well in all but the most demanding failure scenarios, and dual engine failures are generally not considered in aircraft design. Not to be trite, but some things are just too much to consider, like meteor showers and electromagnetic shock waves and so forth. We have to set reasonable design parameters or else nothing would ever be done. This blown tail the way I understand it uses APU air to bolster the boundary layers on TO and landing. There might be some remedy for an APU failure which in this case obviously is more significant than loss of cabin cooling. You could use one of the two remaining engines or there may be some other method that could be used.