Retired USAF here. Some thoughts to consider that may be more fundamental than jumping right into considerations of a portfolio. ( I'm assuming you intend to "fully retire" once you leave the military)
1) Will your expected military retirement meet your "must have/baseline" needs in retirement? If so, then you can tolerate a lot more variability ("risk" in the not-very-accurate terminology often used) in the value of your other investments. This will usually produce higher expected returns. So, you might be fine with 80% stocks/20% bonds (because you can live with very small withdrawals from your portfolio while you wait for the market to return).
2) TSP has very low costs and the unique "G" fund. But the withdrawal options from the TSP are not very flexible, and you won't have access to that money until you reach 59.5 years old (except through a rollover to an IRA and then use of a 72T). So, even if you love the TSP, you'll need some money elsewhere to get you from retirement to age 59.5.
If you don't want to spend much time thinking about this, you could put it all in a Vanguard target-date retirement fund. You won't have to think about it at all, they will automatically rebalance everything for you as the value of various assets changes.
If you want a bit more hands-on, pick any of the moderate "Lazy Portfolios" offered by respected financial writers/blogs. Here's some good coverage/discussion
. My own portfolio is based on Rick Ferri's "core four
" lazy portfolio (also described in the first link) with some added exposure to small stocks and value stocks (VSIAX= Small Cap Value, VTMSX= Small, and some old Vgd Windsor II for value--today I'd probably pick VVIAX--Large Cap Value Index instead). That's really more complicate than it needs to be. I would recommend that you substitute the very good TSP "G Fund" for a portion of the ST bonds in any of these portfolios, since you have access to it. As you have at least 10 years before you'll need the money, and it will need to keep up with inflation for a very long time, you'll probably want an allocation that is heavy in equities (including exposure to intl, small, and value stocks)
The main things: Get a firm grounding in the advantages of low-cost investing and the value of automatic rebalancing (rather then trying to time the market or guess which sector/stock is going to go through the roof). Pick an allocation and leave it alone. If the entire weighted expense ratio of your portfolio is over 0.20%, then (IMO) you are paying too much. There are plenty of great books on this (see the FAQs in this forum). If you are going to be tempted to re-jgger your allocations every time you hear something on CNBC or the market drops 10%, then there's not much point in going down this road.