An appeal to humility might just be an eloquent concession to difficulty. It may not achieve anything if there is something tangible to achieve (for example, your scientific applications). But on the profoundly abstract and inherently human questions it may have a place. In many cases I need to accept that I do not have an answer and will probably never have an answer if I am going to get any sleep at night. But that is a different thing to the 'good' humility which says (a) I am human and capable of making errors and, in fact, it is inevitable that I will err and so accordingly (b) I will implement safeguards against such error in the systems I create and administer. Differing shades of humility appropriate for differing applications?