Can anyone come up with a similar comparison for me? I'd be forever grateful!

The bond length in water (the distance between atomic centers) is a little under one angstrom (Å), or 10^-10 meters. The atomic radius of oxygen is variously found to be from 0.60 Å to 1.52 Å. The covalent radius is 0.73 Å, and that's most commonly used in chemistry, but unfortunately it's defined as half the distance between two singly-bonded oxygen atoms, and the sum of the covalent radii of two bonded atoms is taken to be the bond length! (Since the covalent radius of hydrogen is 0.37 Å, we do in fact find that the predicted bond length in water is 1.00 Å; it's usually taken to be 0.98 Å). So if we use covalent radii, we predict that the distance (normally understood as the distance between "surfaces") between H and O in water is zero. Or, to put it another way, the distance between atomic centers would be about 4 centimeters if the oxygen atom were taken to be about the size of an orange (which I am estimating as 3 centimeters radius).

But perhaps you mean the oxygen nucleus. This is problematic, because atomic nucleii are almost point objects (no dimensions), but we can use the thermal neutron capture cross-section as an approximation of nuclear size. For ¹⁶O, the thermal neutron capture cross-section is about 0.2 millibarns, or 2x10^-28 square centimeters. That works out to a nuclear radius of about 8x10^-15 centimeters, or 8x10^-7 Å.

The problematic nature of taking neutron capture cross sections as nuclear diameters may be seen by the fact that, while ¹⁶O and ¹⁸O have cross sections of about 0.2 mbarn, ¹⁷O has a neutron capture cross section more than 1000 times that, or 235 mbarn! This simply means that ¹⁷O is much more avid for neutrons.

The barn, 10^-24 square centimeters, is an old physics joke ("like hitting the broad side of a barn").