Please explain this statistical concept.?

I'm not sure what you want explained.

A normal distribution curve looks bell-shaped, but not all distribution curves which look bell-shaped are actually normal, though they are likely to be approximately normal.

The bell shape indicates that a large proportion of the observations are grouped around the mean. This is a feature of the normal distribution, with 68.26% of the observations within 1 standard deviation of the mean.

Also the bell shape would indicate symmetry, another important feature of the normal distribution.

The bell shape also tells us the distribution is unimodal, specifically with the mode in the "middle," again being an important feature of the normal distribution.

So to sum up, if it looks bell-shaped it is probably pretty darn close to normal, but a normal curve is actually pretty specific in its conditions. So you can properly say the normal curve are a SUBSET of all probability distribute curves which look bell-shaped.