What does "S.E. of Regression" mean exactly?

[kjd31]

What does this number mean exactly? More importantly, how do you know if it's a good or bad number. If my SE is 6678 - is that remarkably high? Normal? Depends?

I read as long as your SE is less than ~10% of the mean of the dependent variable than it's fine. Further explanation would be great and thanks in advance.

Also - if the Std. Error on the variable coefficient is below .1 does that make it statistically significant at the 10% level or is that the t-stat? Thanks