I disagree - citations count is not a linear measure of quality of work, or importance, and there are significant anomalies and unfairness's in the way things get cited. For example, write an early text book and you may get a tranche of citations without doing a significant amount of important original research. Grad students who co-author with famous authors also get a significant boost due to the reputation and connectivity of their mentor.
But, I think that most senior academics will have less than 15k citations in total. 17k citations for a single paper is remarkable; it's an indication of work that was important, original and definitive.
I only referee in compsci, but in compsci citing papers that are nothing to do with the work in hand, and citing them to generate authority for your arguments does not wash and leads to rejections. Citations point to the basis of your work; establishing that you are building on firm foundations and differentiating your contribution from previous contributions to enable the reader to learn. Being cited in this way is being a teacher to the community.
