All knowledge falls into one of two categories: practical and theoretical. The key to a successful career in computer science is balancing your interest in both.
I consider any topic which is independent of a specific programming language or platform to be theoretical knowledge. Theoretical knowledge encompasses abstract CS topics, such as data structures and algorithms; the underlying mathematical concepts, such as graph theory, combinatorics, and automata; and the hard problems of CS, such as computer vision, artificial intelligence, etc. These are powerful tools for solving hard problems and building things which haven't been possible before.
However, becoming an expert on a theoretical topic won't get you any closer to building a successful iOS app or web service. Without practical knowledge, you'll be dependent on your partner, team, or organization to allow you to accomplish your goals. You won't be able to implement your ideas on your own.
Practical knowledge is the stuff you need to know to bring your idea to the world. It's the connection between abstract ideas and useful products. Practical knowledge includes programming languages, APIs, applications, etc. Mastery of a few practical topics allows you to quickly build something which others can touch and use. Practical knowledge also brings independence: you can start a business if you have something to sell.
However, practical knowledge isn't enough to change the world or even make a competent programmer. Without theoretical knowledge, you will waste time reinventing algorithms and trying to solve impossible problems. In addition, your plans will be bounded by the extent of your theoretical knowledge.
In practice, it's impossible to work in Computer Science or programming without some knowledge from the opposite field. Nevertheless, you can't reach your potential in either field without seeking a balance theoretical and practical knowledge.