Wtf am I doing wrong?, CTA bullets itt

I was working on the four bullets you see on the left here. This is the 5.8mm SIRIPH, an RDX propelled PCTA rifle cartridge meant to have similar properties to 5.56mm NATO that also shares some traits from 6mm optimax.

2 px = 1 mm

When I drawed this, I relied on the outer dimensions for scale. I had previously drawn a magazine before I had drawn the bullet, and a case of this size fitted inside the magazine. So without measuring the inner dimensions (and particularely the bullet dimensions) I drawed the inside of the cartridge. You can guess where this is going.

When comparing my bullet one of the few available pictures of the LSAT bullets (and possibly the only one showing internals) here, I figured the dimensions were about right (well, at second glance the bullet is a little oversized but that doesn't matter). So I drew from this image. (Yeah I know, it's CLA not CTA, please bear with me because from the looks of it they both have pretty much the same dimensions with the exception of the casing.) Then I started bothering with scale again, guess what? 10mm bullet instead of a 5.8mm bullet.

At the far right of the above image you see a (quickly drawn) correction of this. It looks wrong. Like the bullet is either way too small or has way too much propellant behind it.

There's three possibilities that popped up in my mind, and after several headaches I still could not figure out which one is true:
1: The bullet is too short in length
2: A 5.8mm bullet isn't actually 5.8mm in diameter
3: The case is too large
4: An epic clusterfuck over the entire spectrum regarding proportions

Normally when something feels wrong I either ignore it or tweak with it until it looks about right. With ammunition however technical manners such as bullet weight and propellant weight come in play, so I couldn't ignore it any longer. What am I doing wrong?

Rightmost is the only one that doesn't look completely wrong, except you shouldn't place propellant in front of the bullet. Try to guess why.

I thought the cases on the rounds they are testing now are translucent?

e: http://www.youtube.com/watch?v=WlM8IHij6Hs

yep, 0:20, it's not the best picture but you can make out the propellant and the bullet through the case even if it's obscured by the links

Assuming the schematic in the link is to scale, yours doesn't have enough length behind the bullet. If you compare the two, yours is far stubbier.

Well, '5.8mm bullets' are indeed not 5.8mm in diameter. The 5.8mm is the distance between lands in the rifling - the minimum diameter.
The bullet will probably be 6mm due to rifling - like how 5.45mm rounds are 5.62mm in diameter, and that 7.62x51 rounds are sold commercially as .308, not .300 (due to being 7.82mm bullet diameter).

This should be about right.

So I wasn't that far off after all.

That makes sense.

I didn't even notice they were translucent. Although, I don't really see why they should be translucent. I don't see how that would add any benefit.

Thanks for the information, but it would not really matter on this scale; 5.8 is rounded to 12 px which is 6mm.

Danton and Epimachus, thanks. Is this more like it?

You need a cap on the end, probably going at least halfway down the bullet.

There is a cap on the left one. The right one is a cutaway view on which I removed the cap for simplicity. Do I need to extend the cap further?

Ideally it should be long enough to support the bullet, but it seems like you're essentially aiming for a cased caseless round, so I don't know.

It needs to be firmly attached to the case, though, or you'll have tiny plastic bits inside the action.