I'm trying to understand if the average of instantaneous power is equal to real power.
Yes.
My reasoning is that the reactive components average out to zero ...
This is true only if their impedances match and cancel. In your example
$$ Z_C = \frac {1}{2 \pi fC} = \frac {1}{2 \pi \cdot 60 \ 1 \mu} = 2653 \; \Omega $$ and $$ Z_L = 2 \pi f L = 2 \pi \cdot 60 \cdot 1 = 377 \; \Omega $$ so they don't balance. There will be a reactive component in your supply load.

simulate this circuit – Schematic created using CircuitLab
Figure 1. Since the reactive and inductive vectors don't cancel out the result is the kVA vector and it is inductive.
... and from there the average of that signal will give you just the power dissipated by the resistor. Is this true or do we always have to use P=IVpf.
If you are measuring V and I at the supply then you need to use the power factor in your calculations. (If you measure V and I at the resistor then the power factor there will be unity.)