That does not mean that they can't find way to transfer the effect from the mouse to the human, and human trials are different because many other factors are involved, from laws, money, etc bla bla ... With all the differences they are and always will be the best first model to try things on because as i said, with all the differences, there are only 40 genes that are different between humans and a mouse, even if some react differently they still can indicate valuable pathways...
But i agree that they should invest more in human trials but because of the laws in many countries, it's very hard to start them, don't wanna talk about what it means to actually conducted until the end with some results out of it!