これら三つのモデル全ては、高血中鉛濃度（レベル）がより低いテストスコアに関連していることを示している。
といった訳になると思います。で、三つのモデルを、制御変数の組み合わせの正味で見ると、血中鉛濃度との関連がわかる、ということなのではないでしょうか？modelは被験者（動物？ヒト）だと思いますが、test scoreが何のテストのスコアかはわかりません。なのでカタカナ表記にしました。モデルもカタカナです。前後がわからないので何とも言えませんので、netが意味するところもわかりません。ただ、制御変数の一セットのnetが血中鉛濃度の関連性（付随性）を示している、ということだとは思います。

--- I am assuming, and it seems logical to me, that all 3 models used the same control variables that gave the results.

--- Not knowing the rest of your document and what the “three models” look like, I am thinking that your sentence it means:

All three models indicate conclusively (through the process or progression) by a set of control variables that higher blood lead levels are associated with lower test scores.

(with the word "conclusively" I used "by" as in: conclusively by a set of control variables)

My interpretation is also based on the following definitions of the word “net.”
1) “Net” means: Conclusive in a process or progression [Source: WordWeb Dictionary]
And, “conclusive” means: 決定的に, 最終的に, 終結的に

2) “Net” means: 結局の, 最終的な：the net result = 最終結果 [Source: Shogakukan E-J Dictionary]

3) Here is a sample of “using the same control variables in all models”:
Temporal pattern in the effect of postnatal blood lead level on ...
by L Schnaas - 2000 - Cited by 24 - Related articles
We used the same control variables in all models, selected a priori ... We found such ahigh degree of collinearity between HOME scores and ... We examined the diagnostic testsfor general linear models and, .... The unselected group had significantly higher blood lead levels and lower GCI than the selected group. ...
linkinghub.elsevier.com/retrieve/pii/S0892036200