Let’s say there are 10 guys in a bar and the "average" income of them is $40K/yr. Then someone like Bill Gates walks into the bar and has a drink with them. The average pay of those guys would be in the multiples of millions no matter if you use mean, median or mode.
No offense intended but in your example, you are only correct that the mean(average) would skew radically higher because of the presence of Bill Gates in the sample data set. The median, however, is the middlemost number so for the pre-Gates data set that that would have been the #5 placeholder. The mode(the most frequently occurring number in the data set) would remain unchanged.