The author begins with the important first step of self-examination as to one's strengths and weaknesses, career goals and interests. One of my first recommendations to graduate students, postdocs and faculty (even department chairs) is that they read or re-read Managing Oneself by Peter Drucker, published in the Harvard Business Review in 1999 and reprinted in January, 2005. (Search for a pdf version online.) The article is 10 pages and packed with the kinds of insights and questions a good mentor will provide.
Although sciencementor does not (as of this date) discuss the importance and difficulty of learning good time management skills, that is second on my self-mentoring checklist. At the very least, one should know the difference between important and urgent. I recommend that mentees adopt some of the strategies in time-management classics like "Getting Things Done" by David Allen; in particular, email inbox and calendar management are essential for most professionals. My favorite synthesis of self-management is "The 7 Habits of Highly Effective People" by Stephen R. Covey. These resources are not just for business school graduates; I believe that all professionals will be more effective and experience less stress if they learn the skills portrayed in these books.
The self-mentoring blog next describes the formation of an Individual Development Plan: a structured outline of career objectives and action steps to achieve these objectives. My institution, like many others, has adopted a postdoctoral review process that involves the formation and annual review of such a plan; sciencementor provides several excellent templates. These plans also provide the key elements of the postdoctoral mentoring plan now required for NSF grants.
The second section of sciencementor's self-mentoring guide concerns building a professional network. The approaches used here are somewhat field-dependent; in academia, conference attendance and travel to give seminars and colloquiua are more important than career fairs. However, the latter are a good idea too when approaching critical transitions where having alternatives can be helpful. Sciencementor also mentions LinkedIn; while I don't believe it is currently widely used in academic astronomy, the connections available therein can be very helpful in general. LinkedIn is how I learned about sciencementor's self-mentoring blog in the first place!