models with Delta AIC > that of the dot-model

[Andrea]

Hello,
This is a general question about model selection.
In the case where the dot model has a Delta-AIC of around 5, am I correct in thinking that models that include covariates that have Delta-AICs greater than that of the dot-model do not need to be considered for drawing inferences? That is, even if they have a Delta-AIC lower than 7 we don’t need to follow the rule of thumb suggested by Burnham & Anderson 2002 of models with Delta-AIC 0-2 having substantial support, 4-7 having considerably less support and >10 having essentially none?

Andrea