It may not entirely apply to OP's question, but the study [Dion et al.: "Gendered Citation Patterns across Political Science and Social Science Methodology Fields"](https://www.cambridge.org/core/journals/political-analysis/article/gendered-citation-patterns-across-political-science-and-social-science-methodology-fields/5E8E92DB7454BCAE41A912F9E792CBA7) reveals that there may, indeed, be subtle biases in *which* papers get considered foundational or central, and therefore garner more citations. 

As the other answers noted, self-promoting, getting involved in the review process, better networking, and emailing your work to authors when you see a relevant preprint on Arxiv, all might help. In the case of inequality, one could speculate that some authors engage in these activities more than others, or that innate biases mean that other people's prejudices affect the effectiveness of said activities. 

I read a piece recently (can't seem to find it on Google now) that called for removing citation limits and encouraging more comprehensive referencing, to better acknowledge the contributions of young and minority investigators. The implication was that: when the number of references is limited, authors favor works by established (often white-male) colleagues, which might be considered more "canonical", simply due to network effects. 

We might be able to improve this as peer-reviewers: whenever we get a manuscript to review, we should always perform a comprehensive literature search to ensure that the authors are not unintentionally ignoring some papers—especially works by young or minority researchers, which may be high quality but neglected due to bias and network effects.